Every CVE Search/ Get API key

CVE-2019-10099

Sensitive data written to disk unencrypted in Spark

7.5 HighFix available

CVE-2019-10099 is a high vulnerability (CVSS 7.5) in Apache Spark, published August 7, 2019. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 1.30%, the 70th percentile. Fixed in org.apache.spark:spark-core_2.11 2.3.3.

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Attack profile

How easy it is to exploit

Attack vector
Network
Complexity
Low
Privileges needed
None
User interaction
None

What an attacker can affect

Confidentiality
High
Integrity
None
Availability
None

Exploit likelihood

1.30%Down 1.2 points since Apr 14, 2021
90%50%10%1%

What each source reports

NVD Score used 7.5

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
CVE-2019-10099
CVSS 3.1
7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
CVSS 2.0
4.3 MediumAV:N/AC:M/Au:N/C:P/I:N/A:N
Weakness
CWE-312
Updated
2026-11-02
CVE List No score

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
CVE-2019-10099
Updated
2026-11-02
EUVDEUVD-2019-0110 No score

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
EUVD-2019-0110
Updated
2026-11-03
GitHub AdvisoriesGHSA-fp5j-3fpf-mhj5 8.7

Sensitive data written to disk unencrypted in Spark

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
GHSA-fp5j-3fpf-mhj5
CVSS 4.0
8.7 HighCVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N
CVSS 3.0
7.5 HighCVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
Weakness
CWE-312
Updated
2026-11-02
OSV 7.5

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
CVE-2019-10099
CVSS 3.1
7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
Updated
2026-11-02
OSVGHSA-fp5j-3fpf-mhj5 8.7

Sensitive data written to disk unencrypted in Spark

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
GHSA-fp5j-3fpf-mhj5
CVSS 4.0
8.7 HighCVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N
CVSS 3.0
7.5 HighCVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
Weakness
CWE-312
Updated
2026-11-02
OSVPYSEC-2019-114 No score

Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.

Record
PYSEC-2019-114
Updated
2026-11-02

Affected products

Affected packages

Affected packages
PackageAffectedFixed
org.apache.spark:spark-core_2.11
Maven
before 2.3.3
2.3.3
pyspark
PyPI
before 2.3.3
2.3.3
org.apache.spark:spark-core_2.11
Maven
before 2.3.3, plus listed versions
1.2.01.2.11.2.21.3.01.3.1
Show 24 moreShow less
1.4.01.4.11.5.01.5.11.5.21.6.01.6.11.6.21.6.32.0.02.0.0-preview2.0.12.0.22.1.02.1.12.1.22.1.32.2.02.2.12.2.22.2.32.3.02.3.12.3.2
2.3.3
pyspark
PyPI
before 2.3.3, plus listed versions
2.1.12.1.22.1.32.2.02.2.1
Show 5 moreShow less
2.2.22.2.32.3.02.3.12.3.2
2.3.3

History

History
WhenChangedSources
2026-11-03aliases, productsEUVD
2026-11-02createdCVE List, GitHub Advisories, NVD, OSV

References

MoveEnter Open